Publication of the revision of the DSM-III sets the occasion for a series of additional studies on the phenomenology, reliability, validity, and nature of anxiety disorder categories. Building on a series of studies on these topics conducted over the past several years we propose to examine a number of these issues. The Anxiety Disorders Interview Schedule, recently revised to be fully compatible with DSM-III-R will be utilized to determine the reliability of DSM-III-R categories using stringent test - retest methodology. Data will be collected from psychological test, physiological assessment, and periodic self-monitoring of relevant cognitive and behavioral events to ascertain similarities and differences among the DSM-III-R anxiety disorder categories as well as normals and two well-defined stress disorders, irritable bowel syndrome and hypertension. Response to CO2 challenges, and response and habituation to unexpected tones will comprise an important part of the physiological assessment. Patterns of co- morbidity will be determined for the anxiety disorder categories using hierarchy free methods and considering the functional relationships among disorders. In addition, key features of anxiety disorders based on a priori, clinical, or theoretical specification will be identified in terms of their presence or absence, as well as severity across the anxiety disorders. These key features will include panic, social anxiety, anxiety sensitivity, depression, worry, obsessions and compulsions and specific fears. All key features will be measured by questionnaire, interview, and possibly physiological responding. For example, responses to CO2 inhalations may be an index of anxiety sensitivity. Preliminary evidence from the first phase of this project indicates that many of these key features are present in other anxiety disorders (as well as the normal population) and that these features can be dimensionalized to provide profiles within each of the anxiety disorders with substantial implications for classification. Finally, patients in each of the reliably diagnosed DSM-III-R anxiety disorder categories will be followed for a period of two years to determine the course of these disorders and the stability of diagnostic categories and/or dimensional profiles.